🚀 go-pugleaf

  [Dr. David Kahaner is a numerical analyst on sabbatical to the 
   Office of Naval Research-Asia (ONR Asia) in Tokyo from NIST.  The 
   following is the professional opinion of David Kahaner and in no 
   way has the blessing of the US Government or any agency of it.  All 
   information is dated and of limited life time.  This disclaimer should 
   be noted on ANY attribution.]

  [Copies of previous reports written by Kahaner can be obtained using
   anonymous FTP from host cs.arizona.edu, directory japan/kahaner.reports.]

To: Distribution
From: 
 David K. Kahaner
 US Office of Naval Research Asia
 (From outside US):  23-17, 7-chome, Roppongi, Minato-ku, Tokyo 106 Japan
 (From within  US):  Unit 45002, APO AP 96337-0007
  Tel: +81 3 3401-8924, Fax: +81 3 3403-9670
  Email: kahaner@cs.titech.ac.jp
Re: Software patents in Japan
30 Dec 1992
This file is named "patent.92"

ABSTRACT. The changing view of patented information in Japan, with
special emphasis on patenting software.

There is growing concern in Japan about patents, those granted here,
as well as in other countries, especially the US and EC.  Recently, a
variety of publications have focused on disputes, especially with US
companies, and changes in the Japanese patent system. What follows is a
summary of a few of these reports. All technical people should be aware of
the influence that patents and patent policies have on research and
technology directions. Thoughtful comments are welcomed.




Reflecting the growing importance of software in economies and
industries, patents are now rapidly becoming more software oriented in
Japan, Europe, and the US.  Software itself is generally thought to be
protected by copyrights, but how should software "ideas" be protected?

Programs are definitely protected by copyrights.  However, there is no
basis for protecting the ideas and algorithms by copyrights; these can
only be protected by patents.  In the US and Japan the numbers of
software patents, which protect software ideas and methods, are now
increasing rapidly.

Software is incorporated in almost all high-tech devices, especially,
but not limited to, computers.  These are hardware, but software is
essential to their operation.  Patents, which in the past were thought
of as centering on hardware, also protect software ideas and algorithms,
and there has been a rapid advance of patents covering software.

TI and AT&T - Companies that Anticipated the Shift to Software Patents

Anticipating the worldwide shift towards patents covering software, US
firms such as Texas Instruments (TI) and AT&T changed their patent
strategies from an earlier focus on hardware to patent strategies now
centering on software.  There has been a recent drift in this direction
in the patent strategies of other US firms too.

In 1986, TI, which holds numerous basic patents in the field of
semiconductors, brought a lawsuit against nine Japanese and Korean
semiconductor companies for infringing upon its basic DRAM patent.  The
suit was settled out of court.  TI is said to have received more than
$200 million in royalties at that time.  In 1989, TI asked for new
royalty payments from the companies due to the fact that this basic IC
patent, came into existence in Japan.  A lawsuit over this is
still going on (in Japan) between TI and Fujitsu.

AT&T's software patent strategy has been straightforward.  In 1988 AT&T
acquired a patent in the US with Karmarker's linear programming
algorithm (Karmarker, born in India, developed the algorithm while
working at Bell Labs).  An application for the patent was also made in
Japan, but it was rejected although this is now being challenged in court.

In 1989 AT&T acquired a patent for a CIM (computer integrated
manufacturing) technique called the "Product Realization Method (PRM),
essentially an algorithm." An patent application was also made in Japan.
PRM provides computerized centralized control of a series of product
development and production processes, from design to processing and
assembly, in connection with the manufacturing of printed circuit boards
for electronic devices.

In the US, the subjects patents cover have been broadened to include
mathematical solutions, system concepts, and publicly known facts:  The
basic patent for Karmarker's algorithm, and again in CIM, the basic
patent for the FMS (flexible manufacturing system) (obtained in the US
by the UK company, Morins).  The extent to which inventions are
protected with patent rights is tending to expand. For example, the CIM
and FMS are software patents that impact on the machine tool industry.
Japanese machine tool makers, who had been adopting hardware-centered
patent strategies, did not think that there was any possibility of such
software patents being granted.

Software patent strategies that are mainly targeted at the protection of
rights over software ideas and algorithms are not limited to TI and
AT&T; many Japanese believe that these are becoming the basic stance of
most US firms in their patent strategies.  Emphasis is especially being
placed on "basic patents" that pertain to the principles and mechanisms
of new products and technologies.

Basic patents are those granted for technological inventions that spawn
many technologies and new products created on the basis of them. While
Japanese companies have been ahead in terms of the number of patent
applications submitted for almost twenty years, the majority of them
relate to patents for improvements in technology (refinement patents).
In the US there are roughly one million researchers, and about half that
number in Japan. But patents in the US number about 100,000 whereas in
Japan there are over five times that many. The primary reason is that
the subjects of US patents are "basic", whereas Japan's center around
production and refinement technology.  In other words, Japanese lag
behind the US and EC in terms of holding basic patents, and this has
been a major reason for the frequent payment by Japanese companies of
settlements and damages stemming from patent disputes.  Fujitsu's view
is probably close to that generally held in Japan. "New technologies in
electronics are comprised of numerous existing technologies. There are
not any new technologies that do not include other patents." "At the
moment, it would be practically impossible" to develop a basic patent,
such as the one that TI has for basic IC technology and which Japanese
and other companies pay very large sums to use.

A widely held Japanese view is that in the early 1980s, there was a
rapid hollowing out of US manufacturing technology because of the high
value of the US dollar.  During the Reagan administration, strengthening
and protecting intellectual property rights was emphasized, instead of
strengthening manufacturing, as a way to compensate for the decline in
production capacity and to reap profits, and this became a national
policy of the US.  One result was that US firms and the industrial world
significantly shifted the course of their patent strategies from
hardware subjects to an emphasis on software. 

One Japanese writer said simply, "The Minolta...case was inevitable. The
US had been carefully preparing for ten years. Hidden behind that was a
fear of Japanese technology." S.Garai, Chairman of Cannon Corp, which
was involved in a major patent dispute with Honeywell states: "With the
percolation mergers and acquisitions in the US to make money, and of
product liability cases due to excesses, the atmosphere for
straightforward cultivation of manufacturing business has become thin.
In such a situation, I suspect that intellectual property rights have
emerged as a weapon for easily attacking Japan.  That was not
necessarily beneficial in restoring the competitive strength of US
industries.  Nevertheless, as software has become important in economy
and industry, patenting will increasingly tend towards greater emphasis
being placed on software instead of hardware."

"Basic patents, which are given for inventions of basic principles and
mechanisms, can bring huge profits if successful in a single shot, but
there is no guarantee that such original and pioneering inventions will
continually come into existence. It is extremely risky for firms and
national economies to rely on the income received from such basic
patents, or to excessively rely on that growth and development. In
contrast, refinement patents, which recognize inventions that are
improved technology, will necessarily continue to come into existence as
long as a firm has places of manufacturing. Accordingly, only if patents
on inventions that improve quality (not just mass production of poor
articles) keep coming, as they have been up to now [in Japan], will a
firm's continued growth and development be guaranteed. If basic patents
are combined with refinement patents in the right way, Japanese firms
will stand in a very dominant position in the patent war among Japan, US
and EC.  [This author goes on to illustrate the latter point by citing
the web of basic and refinement patents acquired by Xerox during the
1960s on copiers, and that company's subsequent domination of the copier
market for many years.] The current patent strategies of US firms who
have abandoned manufacturing and have over-relied on the acquisition of
basic patents look powerful at a glance, but are fragile in reality.
Japanese firms should absolutely steer clear of the rut that US firms
are now in."

Another Japanese writes, "IBM, a firm that knows the value of
intellectual property much more than AT&T or TI, adopts a realistic
patent strategy that copes with the recent importance of software
patents; it protects its software with patents as well as copyrights.
IBM will never let go of its manufacturing technology and plants that
make hard disks and printers, and in its patent strategy, it regards the
refinement patents born from its manufacturing plants, and not just the
basic patents born from its labs, as important. There is much to learn
from the corporate policies of IBM."

Other Japanese don't agree. For example, a Japanese patent attorney,
F.Otuska, states that viewing patent disputes such as TI's as a conflict
in which US businesses that are losing in the technology competition try
to fight back using patents as weapons, is inappropriate, and that in
fact these examples are few and far between. [On the other hand there
are many cases of non-public reconciliation between companies, so it may
not be possible to accurately judge this trend.]

Otuska states that "Americans feel very strongly about their rights.
Their feeling is beyond comprehension to the average Japanese. It is
easier to understand the situation if we consider most pending patent
disputes [in the US] to have resulted from this American sense of
rights. To protect its own rights, an American firm will fight against
anyone, be it an American or a Japanese firm. [Kodak vs Polaroid, or
NASA vs Hughes are notable examples.]"

"In patent infringement suits, American standards for judging
advancements or improvements achieved by an invention, a requirement in
a patent, are lower than the standards used in Japan. The interpretation
of the scope of a patent right is also expansive. [A good example of
this was a case between Corning and Sumitomo, concerning the
manufacturing of optical fibers. Corning's technology refracts light in
the fiber core, Sumitomo's refracts in the cladding. A US Federal
District Court ruled that these were essentially equivalent.] In Japan,
the effects of an invention are given relatively high weight in judging
advancements achieved by an invention, and the scope of a patent right
tends to be limited to that claimed in the patent application.  Japanese
businesses accustomed to the Japanese patent system are likely to
interpret US patents in the Japanese way. This trend probably made
Japanese businesses misjudge their chances in patent disputes.

"The Japanese patent system contains elements of an industrial
legislation. The US constitution also defines advancing industry as an
objective of protecting inventions, but the US trend to respect the
protection of inventions and ideas is much stronger than the trend in
Japan. We should interpret the US system of rewarding the first inventor
and renewing patents as a manifestation of the US idea of protecting
inventions." Another difference between US and Japanese patent systems
is that the US system  grants patent rights for 17 years from the date
the patent is granted (no matter how long the patent examination
extended). This is linked to the principle of rewarding the first
inventor; the Japanese system awards patent priority to those who apply
first. Japanese patent rights cases use judges; in the US jury trials
are common, 50-60%. Basic opinion in the US is that anything that is
born out of the intelligent activities of human beings is to be
protected (example: "trade-dress"--the yellow color of Kodak's film
cases). This was something that was impossible to understand for
Japanese, who are permitted to borrow upon other's cleverness in all
matters.

Cannon's Garai says, "It is important to develop innovative patents, but
a great deal of time and ideas are necessary before they are implemented
and offered to the consumer at a fair price. While there may be only one
inventor, the corporation invests thousands of people in development and
production,..., a great deal more creativity may be invested in
converting the patent into a product. Since it [the company] made the
patent into something useful, it at least deserves some gratitude."

In the US and Europe, several inventions can be incorporated into a
single application, whether it is for a hardware or software patent.
Even if implementation examples in the application that validate the
invention are few, patent-rights policies have been adopted that grant
"strong, broad rights."  A major factor in granting powerful rights to
the inventor was that it makes it easier to acquire basic patents.

Amidst the increasingly serious patent disputes between Japan, the US,
and EC, the Japanese Patent Office has revamped its
"implementation-example centered doctrine."  This had granted much
narrower rights to the inventor than Patent Offices in the US or EC.
The Japanese are now formulating and implementing new patent examination
criteria that will give "strong, broad rights" like the US and EC, to
original and pioneering inventions.

In the past the Japanese Patent Office's examination policies emphasized
refinement patents because of its "implementation-example centered
doctrine," the new policies aim to encourage firms to also try to
acquire basic patents with "strong, broad rights."  This is a
significant shift in the direction of government patent policies, and
the consequences are also expected to be significant. As readers should
note by some of the quotes in the earlier portion of this report, not
all Japanese think that this approach is correct.

US firms' patent strategies are shifting from hardware to software, as
illustrated by the TI and AT&T examples above.  Moreover, the focus is
narrowing to the acquisition of basic patents with "strong, broad
rights." Japanese feel that the weakening of US manufacturing implies
that refinement patents are not being generated from the manufacturing
environment.  Thus these firms are compelled to narrow the targets of
their patent strategies down to "the acquisition of basic patents in the
software field." [Doesn't imply that in the past US companies did
actually generate refinement patents? Do the statistics support that?]

A Japan view is that the change in the Japanese Patent Office's
examination policies will become a trigger that encourages Japanese
firms to acquire basic patents.  At the same time, though, it will spur
US and EC firms, who already have numerous basic patents and who
emphasize the acquisition of royalties, to apply for and obtain more
patents in Japan.  Given that Japan and the US can cover a huge market
with a single patent application and acquisition, the economic effects
of patents range over an extremely large market. For that reason, the
lawyers of US firms see that "patents become money," and are putting
emphasis on patent offensives against Japan.

"Once the Japanese Patent Office grants "strong, broad rights" with its
adoption of new examination criteria, firms may lose interest in
acquiring many peripheral (refinement) patents that only have narrow,
detailed rights.  Then, the patent strategies will change to the
direction of "competing for quality" more so than volume, and as a
result the number of Japanese patent applications will also decrease."

"Nevertheless, at the same time in Japan there may also be an
increase in the number of patent lawsuits involving the extent to which
the rights of software patents and basic patents are protected, and
patent disputes between Japan, the US, and EC may increase."

In Japan, if "strong, broad rights" are granted with the adoption of new
examination criteria, disputes revolving around the extent to which
rights are protected by patents will be brought into the courts, and
patent lawsuits will increase.

Recently, US firms have been very vigorous in seeking large damage
settlements for patent infringements. A well known example is the
Honeywell vs Minolta case associated with autofocus technology, and also
illustrated the "strong, broad rights" concept. Honeywell obtained
patents from 1975 to 1977. Two images pass through a lens and are then
formed on an optical sensor. If the camera is focused, the two images
will have the same position. Although Honeywell never made a camera, its
technology finds the focusing position by moving the lens with a motor.
Minolta's approach detects  blurring between the images and a
microcomputer calculates the direction in which the lens should be moved
and the amount of lens movement necessary, eventually feeding this to a
motor. A US jury decided that in principle, the technologies are the
same, and that  being able to determine focus in this way (pupil
proportioning) was a theory. Eventually, Minolta made a large
out-of-court settlement with Honeywell.

Several Japanese writers have stated that Japanese firms need to develop
new strategies to confront what is seen as patent offenses by US and EC
companies. These suggestions include the following.

(a) Building new patent strategies that cope with the era of software patents.

(b) Strengthen R&D, especially basic research, in order to emphasize the
acquisition of basic patents that have "strong, broad rights." As
pointed out above, this point is not universally agreed on.

(c) Modifying the usual Japanese routine of paying out large sums of
money to settle patent suits out of court, and thoroughly challenging
the points that should be challenged in the patent trial in order to
check the increase in patent lawsuits and the huge out-of-court
settlements.

Japanese writers are also projecting that patent fights are anticipated
in liquid-crystal and superconductor technology. 

One example of a more aggressive approach, is the Sanyo Electric Co.
Sanyo recently presented a case to the US District Court against TI for
violation of anti-trust laws.  This is Sanyo's response to TI's case
against them, which was presented to the district court, on the grounds
of patent infringement of semiconductors. Sanyo believes that patent
fights are usually conducted whether there exist patent infringements or
not, and contrary to this traditional argument, Sanyo is trying to
counter TI by using anti-trust laws as a weapon.

Another example is Hitachi. In 1981 it established a patent division,
covering everything from software copyrights to trade secrets. This was
elevated to an Intellectual Property Rights Headquarters in 1988 and has
moved into what is considered an offensive, rather than defensive
stance. It has a staff of 330, including a tie-up with a law office in
the US, and 60-70 technical researchers who are specialists in IPR
activities. Other companies have established similar organizations and
claim that they generate profit for their parent corporation.

-----------------------------END OF REPORT----------------------------------